sklearn.preprocessing.robust_scale not available?

When I try to use sklearn.preprocessing.robust_scale I get:

SecurityViolation: 0002 Security Violation(s): Accessing preprocessing.robust_scale raised an AttributeError. No attributes with a similar name were found.
There was a runtime error on line 203.

However, preprocessing.scale works.

Why is robust_scale not available?

4 responses

likely an ancient version of sklearn. Last time I checked only minmax/normalizer/standard scalers were available.

I often run into the same problem with pandas as well.

Since Q seems to use Conda it would be great if we can have custom environments with our choice of versions.

On Quantopian (both the research platform and the backtester/trading API), how does one determine version?

for both
import sklearn as sk
print(sk.version)

(or log.info(sk.version) for algos

Ran into the same thing today, would like to request RobustScaler as well since it's... more robust :).